| Summary: | Unclear meaning of "Ignore pieces of length" | ||
|---|---|---|---|
| Product: | LibreOffice | Reporter: | Tuomas Hietala <tuomas.hietala> |
| Component: | Writer | Assignee: | Heiko Tietze <heiko.tietze> |
| Status: | RESOLVED FIXED | ||
| Severity: | minor | CC: | dgp-mail, heiko.tietze, noelgrandin |
| Priority: | medium | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | All | ||
| Whiteboard: | target:24.8.0 | ||
| Crash report or crash signature: | Regression By: | ||
| Bug Depends on: | |||
| Bug Blocks: | 125674 | ||
|
Description
Tuomas Hietala
2024-04-26 17:48:09 UTC
(In reply to Tuomas Hietala from comment #0) > Description: > The UI string "Ignore _pieces of length:" is somewhat confusing and > contradictory. The following explanation is given in Help: "Set the minimum > number of characters to trigger a valid comparison." > > The UI and Help are at odds here, because it's not clear whether the > selected number of characters is to be ignored (as indicated by the UI > string) or if it's the minimum amount to be taken into account (as indicated > by Help) I agree and couldn't really figure out, what is the current behaviour => NEW > "Pieces of length" should IMO also be reworded for clarity. I'd suggest > replacing the string with something like "Minimum number of characters to > compare" (if Help is correct) or "Maximum number of characters to ignore" > (if Help is not). I support a terminology, that is much clearer, but I don't know, what's the best solution here cc: Design-Team (who should also decide, if we should separate the two issues into two different reports. The property IgnorePieces in officecfg/registry/schema/org/openoffice/Office/Writer.xcs has the description "Specifies whether short pieces of matched text are ignored." IsIgnorePieces() aka m_bIgnorePieces defined in sw/inc/modcfg.hxx is used as CmpOptions.nIgnoreLen = SW_MOD()->IsIgnorePieces() ? SW_MOD()->GetPieceLen() : 0; in sw/source/core/doc/doccomp.cxx ultimately applied in CommonSubseq::IgnoreIsolatedPieces(). Quickly skimming over the code it seems the option controls whether "isolated pieces" (characters) are ignored and up to which length. I guess the wording "pieces" is used because the option is not only applied in case of (compare) "By Character" but also "By word". Noel, you made a lot of changes in the source code. Please enlighten us. Given the source file, it has something to do with comparing documents, other than that, I don't know. Testing with the autotext of lorem, I see a difference in the track changes depending on word vs. character by how the modification is marked. I don't see any effect in the piece length, eg. ibzum vs. ipsum is always marked as either the word is completely different or the particular characters regardless a piece length of 1 or 3. We may just hide the option to solve the UI issue. Presented the case in the ESC and got acceptance to remove the UI access. Heiko Tietze committed a patch related to this issue. It has been pushed to "master": https://git.libreoffice.org/core/commit/6ddb49bc32c534d196ff3913835aceb88cbc8dfc Resolves tdf#160834 - Unclear meaning of "Ignore pieces of length" It will be available in 24.8.0. The patch should be included in the daily builds available at https://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More information about daily builds can be found at: https://wiki.documentfoundation.org/Testing_Daily_Builds Affected users are encouraged to test the fix and report feedback. |